Exploiting Gene-Expression Deconvolution to Probe the Genetics of the Immune System

Area of Science:

• Immunology
• Genetics
• Computational Biology

Background:

• Immune cell subpopulation abundance is influenced by genetic variations.
• Previous studies were limited by low-throughput cell-sorting technologies, focusing on few immune cell types.
• Understanding genetic control of immune cell variation is crucial for immunology and precision medicine.

Purpose of the Study:

• To develop a novel algorithm for identifying the genetic basis of immune cell abundance variation.
• To enable high-throughput analysis of immune cell subpopulations using gene expression and genotyping data.
• To provide a computational tool for dissecting the genetic architecture of immune traits.

Main Methods:

• Developed an algorithm (VoCAL) that predicts immune cell subpopulation abundance from RNA levels of marker genes in complex tissues.
• Integrated predictions from multiple marker gene sets and refined them to mitigate spurious signals.
• Utilized gene expression and genotyping data as primary inputs.
• Implemented the method in the freely available R package ComICS.

Main Results:

• The VoCAL algorithm successfully predicts immune cell abundance and identifies genetic associations.
• Evaluations on synthetic and real biological data demonstrated significant advantages over existing methods.
• The approach allows for the study of a wider range of immune cell types.
• Successfully identified genetic control of predicted immune traits.

Conclusions:

• VoCAL offers a powerful, high-throughput computational approach to investigate the genetic underpinnings of immune cell variation.
• This method advances our ability to study the genetic control of immune cell subpopulations.
• The freely available ComICS R package facilitates broader application in immunological and genetic research.